Look It Up: Bilingual Dictionaries Improve Neural Machine Translation
This addresses translation quality issues for low-resource languages or domains where rare words are problematic, representing an incremental improvement.
The paper tackled the problem of rare words in neural machine translation by developing a method to incorporate bilingual dictionary definitions, resulting in improvements of up to 1.8 BLEU points.
Despite advances in neural machine translation (NMT) quality, rare words continue to be problematic. For humans, the solution to the rare-word problem has long been dictionaries, but dictionaries cannot be straightforwardly incorporated into NMT. In this paper, we describe a new method for "attaching" dictionary definitions to rare words so that the network can learn the best way to use them. We demonstrate improvements of up to 1.8 BLEU using bilingual dictionaries.